In automated machine visions systems, difficulty has been experienced in accurately imaging and inspecting three-dimensional objects and surfaces thereon. This in part is due to the use of typical lenses that distort object characteristics or distances on an object the further away they are from the optical axis. This is further complicated when inspecting surfaces which are not perpendicular to the optical axis. A particularly difficult surface to inspect is the circumferential surface of cylindrically-shaped objects. Many prior techniques have required rotation of the camera about the cylindrical object or alternately rotating the object about its own axis of symmetry to expose the entire surface and taking multiple images thereof.
Thus it would be advantageous to provide a device and method that efficiently and accurately images a three-dimensional object surface and provides an accurate two dimensional or planar image of the entire three-dimensional inspection surface about all 360 degrees with respect to the optical axis.